Early tPA thrombolysis is critical for ischemic stroke treatment, known as time is brain. However, very few patients present for treatment within 3 hours of the stroke onset, the narrow therapeutic window of tPA. Because ischemic tissue injury is heterogeneous, imaging plays a crucial role in stroke patient management. Perfusion and diffusion (PWI and DWI) MRI have proven clinically useful as an imaging approximation to the ischemic tissue prior to infarction (penumbra). The rationale is that PWI identifies hypoperfused tissue while DWI defines the severely damaged ischemic core. As such, the PWI/DWI mismatch identifies the ischemic penumbra, and has been adopted in multiple trials to select patients for tPA therapy. Additionally, variant DWI- based paradigms, including MR angiogram (MRA)/DWI and clinical/DWI have been chosen to overcome the technical challenges of PWI to more practically guide tPA therapy in clinic. However, it is now recognized that the approximation of the DWI lesion as ischemic core is oversimplified. As noted in the recommendation for imaging of acute ischemic stroke from American Heart Association in both 2009 and 2013, DWI is not a simple indicator of irreversible infarction but a complex variable that requires more study. Our proposal to develop kurtosis MRI as a means to augment the standard DWI is thus directly responsive to this call. We have recently established kurtosis MRI for imaging acute stroke. We demonstrated that kurtosis MRI detects the most severely damaged ischemic tissue within the conventional DWI lesion. Importantly, using a transient filament animal stroke model, we showed that kurtosis MRI defines the irreversibly damaged ischemic core while the DWI lesion without kurtosis abnormality recovers upon reperfusion. A key step before we can translate the new kurtosis MRI and guide tPA therapy in acute stroke patients is to evaluate it in an embolic stroke model that more reasonably mimics human stroke. Our central hypothesis is that kurtosis MRI is an ischemic core-specific index, which can overcome the limitation of conventional DWI for imaging acute stroke. Specifically, our proposal will use histology to verify that kurtosis MRI defines more severely injured ischemic tissue (Aim 1), determine kurtosis lesion response to tPA therapy in experimental stroke models (Aim 2), and translate and evaluate kurtosis MRI in the acute stroke clinical setting (Aim 3). The success of our proposal will establish the biologica significance of kurtosis MRI, and establish DKI in the acute stroke clinical setting for future larger scale clinical studies.